The volume trap
Every AI sales cycle (2012's programmatic wave, 2018's sequencer boom, today's agent rush) sells the same promise: more touches per hour. And every cycle, reply rates fall faster than volume rises, because buyers adapt to noise quicker than vendors can manufacture it. Seventeen years of campaign data across our practice shows the pattern without exception: when volume doubles against the same aim, response per touch drops by more than half. The arithmetic of more is negative.
What actually moved the numbers
Three things, in every era, in every industry we measured. First, timing: a message that lands inside the buyer's trigger window outperforms the identical message a month later by multiples, not percentages. Second, reading: the proof-seeker, the price-checker, and the urgency buyer respond to three different first sentences, and one generic opener loses two of the three. Third, persistence with memory: most B2B revenue is lost not to rejection but to silence, in the gap where follow-up died in someone's drafts folder.
Where the new tools fit
Modern models are extraordinary at the parts that were always the bottleneck: reading signals at scale, profiling who is asking, and never letting a thread cool. They are merely adequate at the part everyone uses them for, which is writing more emails. A sales engine built on behavioral reading gets compounding returns from better models; a sequencer built on volume gets cheaper noise.
The test I give every founder
Pull your last twenty lost deals and label each one: rejected, outcompeted, or silent. In most B2B companies, silent wins by a landslide, and silence is an aim and follow-up problem that no volume of new leads fixes. That diagnosis takes one afternoon. The fix is what we build. The math for your industry is in the JSU Bottleneck Index, and a briefing runs your real numbers in thirty minutes.
Request a briefing